EP1886302B1 - Fourniture de synthese de la parole sur des terminaux d'utilisateurs sur un reseau de communications - Google Patents
Fourniture de synthese de la parole sur des terminaux d'utilisateurs sur un reseau de communications Download PDFInfo
- Publication number
- EP1886302B1 EP1886302B1 EP05754668A EP05754668A EP1886302B1 EP 1886302 B1 EP1886302 B1 EP 1886302B1 EP 05754668 A EP05754668 A EP 05754668A EP 05754668 A EP05754668 A EP 05754668A EP 1886302 B1 EP1886302 B1 EP 1886302B1
- Authority
- EP
- European Patent Office
- Prior art keywords
- database
- speech
- user terminal
- incremental
- context
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000015572 biosynthetic process Effects 0.000 title claims abstract description 65
- 238000003786 synthesis reaction Methods 0.000 title claims abstract description 65
- 239000000203 mixture Substances 0.000 claims abstract description 5
- 238000000034 method Methods 0.000 claims description 36
- 238000004590 computer program Methods 0.000 claims description 2
- MQJKPEGWNLWLTK-UHFFFAOYSA-N Dapsone Chemical compound C1=CC(N)=CC=C1S(=O)(=O)C1=CC=C(N)C=C1 MQJKPEGWNLWLTK-UHFFFAOYSA-N 0.000 description 11
- 230000008569 process Effects 0.000 description 9
- 230000007246 mechanism Effects 0.000 description 5
- 230000003068 static effect Effects 0.000 description 5
- 238000013459 approach Methods 0.000 description 4
- 230000001419 dependent effect Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000001308 synthesis method Methods 0.000 description 2
- 241000665848 Isca Species 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L13/00—Speech synthesis; Text to speech systems
- G10L13/02—Methods for producing synthetic speech; Speech synthesisers
- G10L13/04—Details of speech synthesis systems, e.g. synthesiser structure or memory management
- G10L13/047—Architecture of speech synthesisers
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/28—Constructional details of speech recognition systems
- G10L15/30—Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
Definitions
- the present invention relates to a method and to a system for providing speech synthesis on a user terminal over a communications network.
- the present invention relates to a service architecture for providing speech synthesis on user terminals with limited memory availability, such as mobile phones, PDAs (Personal Digital Assistant), personal organizers and digital cameras.
- the invention has been developed with particular attention paid to its possible use in wireless telecommunications networks, for providing enhanced text-to-speech (TTS) services to mobile terminals having embedded a speech synthesizer module based on the concatenation of speech waveforms stored in a database.
- TTS text-to-speech
- Speech synthesis based on concatenation technique is well known in the art, i.e. from patent application WO 00/30069 or from the paper "A concatenative speech synthesis method using context dependent phoneme sequences with variable length as search units", NHK (Nippon Hoso Kyokai; Japan Broadcasting Corp.) Science and Technical Research Laboratories, 5th ISCA Speech Synthesis Workshop, Pittsburgh, USA, June 2004.
- Document WO 00/30069 discloses a speech synthesizer based on concatenation of digitally sampled speech units from a large database.
- a concatenative speech synthesis method using context dependent phoneme sequences with variable length as search units provides a method of dividing an input text into context dependent phoneme sequences and a method of selection of a proper voice waveform database from a static speech database. The speech quality increases when a large speech database is used.
- the Applicant has observed that the quality of such a speech synthesis system, when embedded on a mobile terminal, is intrinsically limited by the maximum database size, which cannot be increased at will on a limited resources terminal.
- Document EP 1471499 Al illustrates a method of distributed speech synthesis, performing a text to speech conversion based on a distributed processing between a remote server and a user terminal.
- the synthesis of speech segments is performed by the server.
- the user terminal downloads synthesized speech segments and concatenates them by means of server rules.
- the user terminal performs a cache mechanism according to the rules provided by the server.
- EP-A1-1431958 discloses downloading different speaker voices.
- Document US 2004/0054534 illustrates an example of speech synthesis customization based on user preferences.
- the user selects voice criteria at a local user terminal.
- the voice criteria represent characteristics that the user desires for a synthesized voice.
- the voice criteria are communicated to a server.
- the server generates a set of synthesized voice rules based on the voice criteria and sends them to the local user terminal.
- the synthesised voice rules represent prosodic aspects of the synthesised voice.
- the Applicant has observed that the speech synthesis quality of above mentioned speech synthesis systems is, as a general rule, directly related to the size of the database of speech waveforms used.
- the Applicant has tackled the problem of obtaining a significant increase in quality of speech synthesis on systems which are embedded on mobile terminals, without affecting too much the memory requirements of the speech waveforms database.
- the Applicant has tackled the problem of dynamically customizing a speech synthesis system based on concatenation technique, achieving the same quality of a static solution based on a database of speech waveforms so huge that it cannot be stored in portable user terminals.
- the Applicant has found that this problem can be solved by providing a network system architecture capable of dynamically downloading incremental databases of speech waveforms and related indexing information that, integrated into the speech synthesis module in the mobile terminal, enhance a reduced database located in the terminal.
- a particular object of the present invention is to provide a service architecture for providing high quality speech synthesis on mobile user terminals with low memory availability.
- the object of the invention is thus to provide satisfactory response to the above needs.
- the dynamic speech synthesis solution herein proposed is able to achieve the same quality of a static solution based on a huge database.
- the dynamic solution works on a virtual huge database by using a reduced static database located into the user terminal and downloading the suitable incremental databases only when it is needed.
- Each incremental database containing further speech elements with respect to the default database, is related to the context of the specific information to be synthesized.
- the extended database given by the default database plus the downloaded incremental database(s)
- the speech synthesis on the user terminal achieves an improved quality.
- Each dynamic incremental database is related to one or more information contexts, e.g. economics, sport, comics and so on.
- the same incremental database can be used to improve the synthesis of several contents concerning the same context.
- a cache mechanism may be implemented into the user terminal in order to avoid further download of the same incremental database when the information context is not changed, or when a same context is used frequently.
- incremental databases can be used for customising the speech synthesis performed on a user terminal with respect to server context rules, to the specific user profile or to specific user preferences.
- a service architecture for providing, on a user terminal 2 having at least one connection (wire line or wireless) with a telecommunication network 12, speech synthesis from textual information comprises, besides terminal 2 and network 12, a content server 16 for downloading contents into the user terminal, a context server 14 for selecting a context associated to the information requested by the user, and a database server 8 containing a plurality of incremental databases DB1, .., DBN to be downloaded into the user terminal 2.
- User terminal 2 is equipped with an embedded speech synthesizer 4 and with a limited default speech database 6. By using this basic configuration, the terminal 2 is able to synthesize any textual information related to any context at a minimum quality level. Textual information may be already present on the user terminal 2 or is retrieved from the remote content server 16 by means of a browser application 18 installed on the same terminal 2 and visualized on a display 20.
- the proposed architecture employs a database server 8 (or a plurality of database servers) able to provide a plurality of incremental databases DB1, .., DBN with the aim of increase the default database 6 located in the user terminal 2.
- a caching mechanism is furthermore provided in the user terminal 2 in order to efficiently manage the download and storage of the incremental databases into the user terminal.
- a cache memory 10 makes available the stored incremental database for all the time in which the relative context is used, as several information required by the user could refer to the same context and therefore to the same incremental database. In this case, no other incremental database is downloaded from the network, the speech synthesis process is accelerated and network resources are optimised.
- the caching mechanism associates a "time to live” parameter to each stored incremental database in order to avoid further downloads when the relative context is reused frequently.
- Parameters governing the caching mechanism such as for example the cache size and the "time to live", can be either set on the user terminal, by user setting, or according to rules provided by the context server 14.
- the user terminal browser 18 sends to the context server 14 a request (arrow 30) related to the information to be synthesized, through the network connection 12.
- a context manager 22 located into the context server 14 asks the content server 16 for downloading the information contents to the user terminal browser (arrow 34) and extracts context information from the information contents.
- the context manager 22 checks if the user terminal has already in cache 10 the suitable incremental database. Several alternatives are possible in order to perform this check, for instance the terminal 2 can communicate the description of already available data bases together with the request (arrow 30), or the context manager 22 can keep track of all previous database downloads 32 associated to a specific user terminal.
- the context manager 22 activates a context selector 24 in order to choose a proper incremental database DB2 in the database server 8, and transfers such incremental database to a database manager 26 on the user terminal 2 (arrow 32).
- Content server 16, database server 8 and context server 14 can be located in a same server or, indifferently, in separate servers connected to a common telecommunication network (e.g. the Internet).
- the database manager 26 on the user terminal 2 receives the incremental database from the network and manages the composition of a new enlarged database, as explained in detail hereinafter.
- the database manager 26 notifies the speech synthesizer 4 that a new speech waveforms database is available, achieving therefore a higher synthesis quality.
- Such relationship can be established at "server side”, i.e. in the context server 14, or at "client side", i.e. according to user preferences.
- server side i.e. in the context server 14, or at "client side”, i.e. according to user preferences.
- client side i.e. according to user preferences.
- server context rules such as economic context, sport context, comics context and so on
- specific user profile stored in the server.
- the decision is performed according to specific user preferences, for example prosody customization, dialectal inflection and so on.
- the basic database 2 of speech waveforms can be replaced by a substitute one, by downloading into the user terminal the new database.
- the downloading and replacement process is managed by the database manager 26, which retrieves the new database from a remote server (e.g. the same database server 8) and replaces the database in the terminal's memory.
- a remote server e.g. the same database server 8
- Such a replacement can be useful, for example, if a user wishes to change the language of the TTS system.
- the speech synthesis process embedded on user terminal 2 can be carried out according to two different techniques.
- speech synthesis can start as soon as the content is available on the user terminal, avoiding further delays, using the default database, providing the minimum quality level.
- the speech synthesis quality is then improved when the incremental database is fully downloaded and merged with the database already present in the terminal, allowing the text-to-speech engine using a larger speech database.
- the speech synthesis doesn't start immediately using the default database, but starts only as soon as the incremental database is available on the user terminal. In this case the speech synthesis is subject to an initial time delay, but the synthesis quality is, from the beginning, constant at high levels.
- FIG. 2 illustrates a process according to the first technique, focusing on parallelism of different processes.
- a user request (step 40) on the terminal browser activates (step 42) the context manager 22 in the context server 14, starting two parallel processes (steps 44, 46) respectively for downloading requested information contents into the user terminal and for checking whether the associated incremental database is already present in the user terminal cache 10.
- the speech synthesis starts (step 52), the same happens if the associated incremental database is already present in the cache 10.
- the associated incremental database is not present in cache, it is downloaded into the user terminal (step 48) and merged with the current database in a new enlarged database (step 50), so that the speech synthesis continues (step 52) with improved quality.
- FIG. 3 illustrates a process according to the second technique not in accordance with the invention, focusing on serialisation of different processes.
- a user request (step 60) on the terminal browser activates (step 62) the context manager 22 in the context server 14 for downloading the requested information contents into the user terminal (step 64) and for subsequently checking (step 66) whether the associated incremental database is already present in the user terminal cache 10. If the associated incremental database is already present the speech synthesis is started (step 72), otherwise it is downloaded into the user terminal (step 68) and merged with the current database in a new enlarged database (step 70), and then started the speech synthesis (step 72). According to this technique the speech synthesis process is delayed until the new enlarged database is available, so that the quality of the speech is constant during synthesis phase.
- Figure 4 illustrates a speech database structure of a speech synthesis system, in particular the default database 6 of the user terminal 2.
- the speech database comprises a database of speech units, or speech waveforms, 80 and an associated descriptor file 82 related to that database.
- the descriptor file 82 contains a reference table that lets the speech synthesizer 4 know the position of the required speech unit in the database of speech units 80.
- the descriptor file 82 is a diphone reference table, where a "diphone" is intended as a fundamental speech unit composed of two adjacent half-phones.
- the diphone reference table 82 contains information about all the diphones present in the database of speech units 80.
- Each diphone #1 .. #N has a corresponding portion E1 .. EN of the table 82, as shown in Figure 4 , containing:
- Figure 5 illustrates a speech database structure after the downloading of an incremental database 92 of speech units over a previous database of speech units 90 already present in the user terminal, and the corresponding upgrade of the diphone reference table 94.
- the upgrade of the diphone reference table 94 comprises, for example, if the downloaded incremental database of speech units 92 contains a speech segment with the diphone #1, integrating the portion E1 of the table relative to diphone #1 with further pointers P'(#1) to those added speech segments.
- the context manager 22 sends to the user terminal 2 a suitable incremental database with an associated new descriptor file (the complete diphone descriptor file 94 in the above example).
- the database manager 26 stores the incremental database in the cache memory 10 and replaces the descriptor file with the new one.
- the context manager 22 sends to the user terminal 2 the incremental database and a patch file that the database manager 26 applies to the descriptor file already present in the user terminal in order to upgrade it to the new descriptor file 94.
- speech synthesis can be performed on a mobile terminal without any network connection, improving usability of the applications/services based on speech synthesis.
- a fully operational TTS engine is in fact always available on the terminal, thanks to the embedded default speech database.
- the speech synthesis quality can be improved, achieving the same high quality of a static solution based on a huge database that could not be stored in a limited memory user terminal.
- service providers can differentiate services based on speech synthesis through the differentiation of the quality of supplied services, according to the availability of proprietary incremental databases.
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Telephonic Communication Services (AREA)
- Mobile Radio Communication Systems (AREA)
- Machine Translation (AREA)
Claims (21)
- Procédé de génération de paroles synthétisées à partir d'informations textuelles sur un terminal d'utilisateur (2), ledit terminal d'utilisateur étant muni d'un moteur de synthèse de paroles (4) ayant une base de données de base (6) de formes d'ondes de paroles, comprenant les étapes consistant à :- commencer (52) la synthèse de paroles desdites informations textuelles en employant ladite base de données de base (6) de formes d'ondes de paroles ,- extraire (42) des informations de contexte associées aux dites informations textuelles ,- sélectionner (42) une base de données incrémentielle (DB1, ..., DBN) de formes d'ondes de paroles associées aux dites informations de contexte ,- fournir (48) au dit terminal d'utilisateur (2) ladite base de données incrémentielle (DB1, ..., DBN) ,- gérer (50) la composition d'une base de données élargie de formes d'ondes de paroles comprenant ladite base de données de base (6) et lesdites bases de données incrémentielles de formes d'ondes de paroles, et- continuer (52) la synthèse de paroles desdites informations textuelles en employant ladite base de données élargie de formes d'ondes de paroles.
- Procédé selon la revendication 1, dans lequel ladite étape consistant à fournir audit terminal d'utilisateur ladite base de données incrémentielle comprend les étapes consistant à :- vérifier (46) si ladite base de données incrémentielle est déjà présente dans ledit terminal d'utilisateur , et- télécharger (48), à travers un réseau de communication (12), ladite base de données incrémentielle dans ledit terminal d'utilisateur, si elle n'y est pas déjà présente.
- Procédé selon la revendication 2, dans lequel ladite étape consistant à télécharger ladite base de données incrémentielle dans ledit terminal d'utilisateur comprend les étapes consistant à :- télécharger une base de données incrémentielle de formes d'ondes de paroles (92) , et- mettre à jour un fichier descripteur (94) de ladite base de données élargie de formes d'ondes de paroles.
- Procédé selon la revendication 1, comprenant en outre l'étape consistant à stocker ladite base de données incrémentielle dans une mémoire cache sur ledit terminal d'utilisateur.
- Procédé selon la revendication 1, dans lequel ladite étape consistant à sélectionner (42) une base de données incrémentielle (DB1, ..., DBN) de formes d'ondes de paroles associées aux dites informations de contexte est effectuée selon des règles de contexte indépendantes de profils d'utilisateurs.
- Procédé selon la revendication 1, dans lequel ladite étape consistant à sélectionner (42) une base de données incrémentielle (DB1, ..., DBN) de formes d'ondes de paroles associées aux dites informations de contexte est effectuée selon des règles de contexte basées sur un profil d'utilisateur.
- Procédé selon la revendication 1, dans lequel lesdites informations textuelles sont extraites par ledit terminal d'utilisateur à travers un réseau de communication (12).
- Procédé selon la revendication 1, comprenant l'étape consistant à :- remplacer ladite base de données de base (6) de formes d'ondes de paroles en téléchargeant, à travers un réseau de communication (12), une base de données de base de substitution dans ledit terminal d'utilisateur (2).
- Procédé selon l'une quelconque des revendications 2, 3, 7 et 8, dans lequel ledit réseau de communication est un réseau de télécommunications sans fil.
- Architecture de services comprenant un terminal d'utilisateur pour fournir au dit terminal d'utilisateur (2) une synthèse de paroles relative à des informations textuelles, ledit terminal d'utilisateur étant muni d'un moteur de synthèse de paroles (4) et d'une base de données de base de formes d'ondes de paroles (6), caractérisée en ce qu'elle comprend :- un serveur de contexte (14) pour télécharger une base de données incrémentielle (DB1, ..., DBN) de formes d'ondes de paroles dans ledit terminal d'utilisateur, la base de données incrémentielle étant associée à des informations de contexte associées aux dites informations textuelles ,- un gestionnaire de base de données (26) sur ledit terminal d'utilisateur (2) pour gérer la composition d'une base de données élargie de formes d'ondes de paroles pour ledit moteur de synthèse de paroles comprenant ladite base de données de base (6) et lesdites bases de données incrémentielles (DB1, ..., DBN) de formes d'ondes de paroles, ledit moteur de synthèse de paroles (4) étant opérationnel pour commencer la synthèse de paroles desdites informations textuelles en employant ladite base de données de base de formes d'ondes de paroles (6) aussi longtemps que ladite base de données élargie de formes d'ondes de paroles n'est pas encore disponible.
- Architecture de services selon la revendication 10, dans laquelle ledit serveur de contexte (14) comprend :- un gestionnaire de contexte (22) pour extraire des informations de contexte desdites informations textuelles ,- un sélecteur de contexte (24) pour sélectionner une base de données incrémentielle (DB1, ..., DBN) de formes d'ondes de paroles associées aux dites informations de contexte et pour télécharger ladite base de données incrémentielle dans ledit terminal d'utilisateur.
- Architecture de services selon la revendication 10, comprenant en outre sur ledit terminal d'utilisateur (2) une mémoire cache (10) pour stocker provisoirement au moins une base de données incrémentielle téléchargée.
- Architecture de services selon la revendication 11, comprenant en outre un serveur de base de données (8) stockant une pluralité de bases de données incrémentielles (DB1, ..., DBN) sélectionnées par ledit sélecteur de contexte (24).
- Architecture de services selon la revendication 11, dans laquelle ledit sélecteur de contexte (24) fonctionne en fonction de règles de contexte indépendantes de profils d'utilisateurs.
- Architecture de services selon la revendication 11, dans laquelle ledit sélecteur de contexte (24) fonctionne en fonction de règles de contexte basées sur un profil d'utilisateur.
- Architecture de services selon la revendication 10, comprenant en outre un serveur de contenus (16) pour télécharger lesdites informations textuelles dans ledit terminal d'utilisateur.
- Architecture de services selon l'une quelconque des revendications 10 à 16, dans laquelle ledit réseau de communication est un réseau de télécommunications sans fil.
- Terminal d'utilisateur (2) pour un réseau de communication comprenant :- une base de données de base (6) de formes d'ondes de paroles ,- un moyen (18) pour extraire des informations textuelles à convertir en paroles synthétisées ,caractérisé en ce qu'il comprend :- un gestionnaire de bases de données (26) pour télécharger une base de données incrémentielle de formes d'ondes de paroles (DB1, ..., DBN) d'un serveur de base de données (8), ladite base de données incrémentielle étant associée à des informations de contexte associées aux dites informations textuelles, et pour gérer la composition d'une base de données élargie de formes d'ondes de paroles pour ledit moteur de synthèse de paroles (4), comprenant ladite base de données de base (6) et lesdites base de données incrémentielles de formes d'ondes de paroles ,- un moteur de synthèse de paroles (4) opérationnel pour commencer la synthèse de paroles desdites informations textuelles en employant ladite base de données de base de formes d'ondes de paroles (6) aussi longtemps que ladite base de données élargie de formes d'ondes de paroles n'est pas encore disponible.
- Terminal d'utilisateur selon la revendication 18, comprenant en outre une mémoire cache (10) pour stocker provisoirement au moins une base de données incrémentielle téléchargée.
- Terminal d'utilisateur selon la revendication 18, dans lequel ledit moteur de synthèse de paroles (4) est basé sur une synthèse de paroles par concaténation.
- Produit de programme informatique, pouvant être chargé dans la mémoire d'au moins un ordinateur et comprenant des parties de code logiciel pour effectuer le procédé selon l'une quelconque des revendications 1 à 9.
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/EP2005/005818 WO2006128480A1 (fr) | 2005-05-31 | 2005-05-31 | Procede et systeme de fourniture de synthese de la parole sur des terminaux d'utilisateurs sur un reseau de communications |
Publications (2)
Publication Number | Publication Date |
---|---|
EP1886302A1 EP1886302A1 (fr) | 2008-02-13 |
EP1886302B1 true EP1886302B1 (fr) | 2009-11-18 |
Family
ID=35252393
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP05754668A Active EP1886302B1 (fr) | 2005-05-31 | 2005-05-31 | Fourniture de synthese de la parole sur des terminaux d'utilisateurs sur un reseau de communications |
Country Status (6)
Country | Link |
---|---|
US (1) | US8583437B2 (fr) |
EP (1) | EP1886302B1 (fr) |
AT (1) | ATE449399T1 (fr) |
DE (1) | DE602005017829D1 (fr) |
ES (1) | ES2336686T3 (fr) |
WO (1) | WO2006128480A1 (fr) |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8600753B1 (en) * | 2005-12-30 | 2013-12-03 | At&T Intellectual Property Ii, L.P. | Method and apparatus for combining text to speech and recorded prompts |
JP2009265279A (ja) * | 2008-04-23 | 2009-11-12 | Sony Ericsson Mobilecommunications Japan Inc | 音声合成装置、音声合成方法、音声合成プログラム、携帯情報端末、および音声合成システム |
CN101593516B (zh) | 2008-05-28 | 2011-08-24 | 国际商业机器公司 | 语音合成的方法和系统 |
PL401347A1 (pl) * | 2012-10-25 | 2014-04-28 | Ivona Software Spółka Z Ograniczoną Odpowiedzialnością | Spójny interfejs do lokalnej i oddalonej syntezy mowy |
US9159314B2 (en) * | 2013-01-14 | 2015-10-13 | Amazon Technologies, Inc. | Distributed speech unit inventory for TTS systems |
CN110460654A (zh) * | 2019-02-26 | 2019-11-15 | 北京蓦然认知科技有限公司 | 一种智能交互平台中交互引擎的下载方法及装置 |
CN111429878B (zh) * | 2020-03-11 | 2023-05-26 | 云知声智能科技股份有限公司 | 一种自适应语音合成方法及装置 |
Family Cites Families (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6173250B1 (en) * | 1998-06-03 | 2001-01-09 | At&T Corporation | Apparatus and method for speech-text-transmit communication over data networks |
US6446040B1 (en) * | 1998-06-17 | 2002-09-03 | Yahoo! Inc. | Intelligent text-to-speech synthesis |
US6665641B1 (en) | 1998-11-13 | 2003-12-16 | Scansoft, Inc. | Speech synthesis using concatenation of speech waveforms |
US6408272B1 (en) * | 1999-04-12 | 2002-06-18 | General Magic, Inc. | Distributed voice user interface |
FI115868B (fi) * | 2000-06-30 | 2005-07-29 | Nokia Corp | Puhesynteesi |
US20020156630A1 (en) * | 2001-03-02 | 2002-10-24 | Kazunori Hayashi | Reading system and information terminal |
US7035794B2 (en) * | 2001-03-30 | 2006-04-25 | Intel Corporation | Compressing and using a concatenative speech database in text-to-speech systems |
US6950798B1 (en) * | 2001-04-13 | 2005-09-27 | At&T Corp. | Employing speech models in concatenative speech synthesis |
JP2002358092A (ja) * | 2001-06-01 | 2002-12-13 | Sony Corp | 音声合成システム |
JP2002366186A (ja) * | 2001-06-11 | 2002-12-20 | Hitachi Ltd | 音声合成方法及びそれを実施する音声合成装置 |
US20030028377A1 (en) * | 2001-07-31 | 2003-02-06 | Noyes Albert W. | Method and device for synthesizing and distributing voice types for voice-enabled devices |
JP3748064B2 (ja) * | 2002-02-08 | 2006-02-22 | 日本電信電話株式会社 | 音声合成方法、音声合成装置および音声合成プログラム |
US20040054534A1 (en) * | 2002-09-13 | 2004-03-18 | Junqua Jean-Claude | Client-server voice customization |
AU2003260854A1 (en) | 2002-10-04 | 2004-04-23 | Koninklijke Philips Electronics N.V. | Speech synthesis apparatus with personalized speech segments |
EP1431958B1 (fr) | 2002-12-16 | 2018-07-18 | Sony Mobile Communications Inc. | Un appareil connectable ou comportant un dispositif de génération de la parole, et programme d'ordinateur adapté |
US7533023B2 (en) * | 2003-02-12 | 2009-05-12 | Panasonic Corporation | Intermediary speech processor in network environments transforming customized speech parameters |
EP1471499B1 (fr) | 2003-04-25 | 2014-10-01 | Alcatel Lucent | Procédé de la synthèse de parole répartie |
US7260539B2 (en) * | 2003-04-25 | 2007-08-21 | At&T Corp. | System for low-latency animation of talking heads |
GB0323551D0 (en) | 2003-10-08 | 2003-11-12 | Radioscape Ltd | DAB radio system with voiced control feedback |
US8666746B2 (en) * | 2004-05-13 | 2014-03-04 | At&T Intellectual Property Ii, L.P. | System and method for generating customized text-to-speech voices |
US7599838B2 (en) * | 2004-09-01 | 2009-10-06 | Sap Aktiengesellschaft | Speech animation with behavioral contexts for application scenarios |
-
2005
- 2005-05-31 EP EP05754668A patent/EP1886302B1/fr active Active
- 2005-05-31 AT AT05754668T patent/ATE449399T1/de not_active IP Right Cessation
- 2005-05-31 DE DE602005017829T patent/DE602005017829D1/de active Active
- 2005-05-31 WO PCT/EP2005/005818 patent/WO2006128480A1/fr active Application Filing
- 2005-05-31 US US11/921,403 patent/US8583437B2/en active Active
- 2005-05-31 ES ES05754668T patent/ES2336686T3/es active Active
Also Published As
Publication number | Publication date |
---|---|
US20090306986A1 (en) | 2009-12-10 |
ATE449399T1 (de) | 2009-12-15 |
DE602005017829D1 (de) | 2009-12-31 |
US8583437B2 (en) | 2013-11-12 |
WO2006128480A1 (fr) | 2006-12-07 |
EP1886302A1 (fr) | 2008-02-13 |
ES2336686T3 (es) | 2010-04-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP1886302B1 (fr) | Fourniture de synthese de la parole sur des terminaux d'utilisateurs sur un reseau de communications | |
JP4130190B2 (ja) | 音声合成システム | |
KR101055045B1 (ko) | 음성 합성 방법 및 시스템 | |
US7693719B2 (en) | Providing personalized voice font for text-to-speech applications | |
US7966186B2 (en) | System and method for blending synthetic voices | |
US7689421B2 (en) | Voice persona service for embedding text-to-speech features into software programs | |
US20190088249A1 (en) | System and method for distributed voice models across cloud and device for embedded text-to-speech | |
KR101076202B1 (ko) | 음성 합성 장치, 음성 합성 방법 및 프로그램이 기록된 기록 매체 | |
US20040099126A1 (en) | Interchange format of voice data in music file | |
CN1675681A (zh) | 客户机-服务器语音定制 | |
KR20070048118A (ko) | 합성 데이터 음성 렌더링의 동적 운율 조정 방법, 시스템및 컴퓨터 프로그램 제품 | |
WO2006106182A1 (fr) | Amelioration de l'utilisation de la memoire dans un synthetiseur texte-parole | |
JP3884856B2 (ja) | 音声合成用データ作成装置、音声合成装置及びそれらの方法、コンピュータ可読メモリ | |
JP2014021136A (ja) | 音声合成システム | |
US5848390A (en) | Speech synthesis system and its method | |
JP3748064B2 (ja) | 音声合成方法、音声合成装置および音声合成プログラム | |
JP4653572B2 (ja) | クライアント端末、音声合成情報処理サーバ、クライアント端末プログラム、音声合成情報処理プログラム | |
EP1665229B1 (fr) | Synthese vocale | |
JP2005018037A (ja) | 音声合成装置、音声合成方法及びプログラム | |
JP4170819B2 (ja) | 音声合成方法及びその装置並びにそのコンピュータプログラム及びそれを記憶した情報記憶媒体 | |
JP2005164701A (ja) | 演奏制御データ生成装置およびプログラム | |
JP2005018036A (ja) | 音声合成装置、音声合成方法及びプログラム | |
JP2003029774A (ja) | 音声波形辞書配信システム、音声波形辞書作成装置、及び音声合成端末装置 | |
EP1831869A2 (fr) | Procede et appareil pour l'amelioration de la performance de conversion texte-voix | |
KR20030037403A (ko) | 음성 합성기술을 이용한 휴대통신 단말기의 벨소리제공시스템 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
17P | Request for examination filed |
Effective date: 20071128 |
|
AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IS IT LI LT LU MC NL PL PT RO SE SI SK TR |
|
17Q | First examination report despatched |
Effective date: 20080519 |
|
DAX | Request for extension of the european patent (deleted) | ||
GRAP | Despatch of communication of intention to grant a patent |
Free format text: ORIGINAL CODE: EPIDOSNIGR1 |
|
RTI1 | Title (correction) |
Free format text: PROVIDING SPEECH SYNTHESIS ON USER TERMINALS OVER A COMMUNICATIONS NETWORK |
|
GRAS | Grant fee paid |
Free format text: ORIGINAL CODE: EPIDOSNIGR3 |
|
GRAA | (expected) grant |
Free format text: ORIGINAL CODE: 0009210 |
|
AK | Designated contracting states |
Kind code of ref document: B1 Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IS IT LI LT LU MC NL PL PT RO SE SI SK TR |
|
REG | Reference to a national code |
Ref country code: GB Ref legal event code: FG4D |
|
REG | Reference to a national code |
Ref country code: CH Ref legal event code: EP |
|
REG | Reference to a national code |
Ref country code: IE Ref legal event code: FG4D |
|
REF | Corresponds to: |
Ref document number: 602005017829 Country of ref document: DE Date of ref document: 20091231 Kind code of ref document: P |
|
REG | Reference to a national code |
Ref country code: NL Ref legal event code: VDEP Effective date: 20091118 |
|
REG | Reference to a national code |
Ref country code: ES Ref legal event code: FG2A Ref document number: 2336686 Country of ref document: ES Kind code of ref document: T3 |
|
LTIE | Lt: invalidation of european patent or patent extension |
Effective date: 20091118 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: SE Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20091118 Ref country code: LT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20091118 Ref country code: PT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20100318 Ref country code: IS Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20100318 Ref country code: FI Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20091118 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: CY Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20091118 Ref country code: SI Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20091118 Ref country code: PL Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20091118 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: AT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20091118 Ref country code: BE Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20091118 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: DK Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20091118 Ref country code: NL Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20091118 Ref country code: BG Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20100218 Ref country code: RO Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20091118 Ref country code: EE Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20091118 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: CZ Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20091118 Ref country code: SK Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20091118 |
|
PLBE | No opposition filed within time limit |
Free format text: ORIGINAL CODE: 0009261 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: NO OPPOSITION FILED WITHIN TIME LIMIT |
|
26N | No opposition filed |
Effective date: 20100819 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: GR Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20100219 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: MC Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20100531 |
|
REG | Reference to a national code |
Ref country code: CH Ref legal event code: PL |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: CH Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20100531 Ref country code: LI Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20100531 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: IE Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20100531 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: HU Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20100519 Ref country code: LU Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20100531 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: TR Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20091118 |
|
REG | Reference to a national code |
Ref country code: FR Ref legal event code: PLFP Year of fee payment: 12 |
|
REG | Reference to a national code |
Ref country code: FR Ref legal event code: PLFP Year of fee payment: 13 |
|
REG | Reference to a national code |
Ref country code: FR Ref legal event code: PLFP Year of fee payment: 14 |
|
P01 | Opt-out of the competence of the unified patent court (upc) registered |
Effective date: 20230529 |
|
P02 | Opt-out of the competence of the unified patent court (upc) changed |
Effective date: 20230602 |
|
PGFP | Annual fee paid to national office [announced via postgrant information from national office to epo] |
Ref country code: IT Payment date: 20230519 Year of fee payment: 19 Ref country code: FR Payment date: 20230525 Year of fee payment: 19 Ref country code: ES Payment date: 20230601 Year of fee payment: 19 Ref country code: DE Payment date: 20230530 Year of fee payment: 19 |
|
PGFP | Annual fee paid to national office [announced via postgrant information from national office to epo] |
Ref country code: GB Payment date: 20230529 Year of fee payment: 19 |